Annals of Hepatology 2008; 7(2): April-June: 148-151
Hepatology Impact of diabetes mellitus on outcome of HCC
Deepak N. Amarapurkar;1 Nikhil D. Patel;2 Praful M. Kamani2
Abstract Key words: Hepatocellular carcinoma, Diabetes Mellitus,
Non-alcoholic fatty liver disease, Chronic liver disease.
Background: Diabetes mellitus (DM) is recently identi-
fied risk factor for development and progression of Introduction
chronic liver disease as well as hepatocellular carcino-
ma (HCC). We planned a prospective analysis to identi- Type-2 diabetes mellitus (DM) increases risk of develop-
fy impact of DM in Indian patients with HCC. Methods: ment of chronic liver disease (CLD).1 Commonest CLD in
During last 10 years, 160 consecutive patients of HCC DM is non-alcoholic fatty liver disease (NAFLD).2 Hepato-
were evaluated. Demographic profile like age of presen- cellular carcinoma (HCC) occurs in patients with CLD and
tation, clinical features, etiology of HCC, tumor size at mostly in presence of cirrhosis. Known predisposing causes
presentation, management and ultimate outcome was of HCC are hepatitis B virus, hepatitis C virus (HCV),
compared diabetic with non-diabetic HCC patients. Re- chronic alcohol abuse, hemochromatosis and, recently, non-
sults: During last 10 years, 160 consecutive patients of alcoholic fatty liver disease (NAFLD).3 Cryptogenic cirrho-
HCC were evaluated (Mean age = 59.6 ± 12.9 years, sex sis remains responsible for HCC in 15-50% cases.4,5
ratio (M: F) = 5.4: 1). Etiology for HCC were hepatitis B Although earlier studies denied the association be-
in 45 (28.2%), hepatitis C in 18 (11.3%), alcohol in 27 tween type-2 diabetes mellitus (DM) and HCC;6-8 in most
(16.8%), alcohol with hepatitis B in 12 (7.5%), alcohol of the recent studies, DM is shown to increase risk of HCC
with hepatitis C in 1 (0.6%), non-alcoholic steatohepati- by 2- to 4-fold, even after adjusting for other predisposing
tis in 4 (2.5%) and cryptogenic in 53 (33.2%) patients. factors.5,9-29 In presence of viral hepatitis and alcohol in-
Patients of HCC with DM (group-A, n =46, age = 62.6 ± take, DM increases risk for HCC by 10-fold.25 DM is a ma-
9.5 years, sex (M: F) = 6.6:1) were compared with pa- jor risk factor for NAFLD,30-33 which is shown to predis-
tient of HCC without DM (group-B, n =114, age = 66.7 ± pose for cryptogenic cirrhosis34 and HCC.35-37 There is in-
13.7 years, sex (M: F) = 5.4:1). Duration of diabetes in creased incidence of DM in HCV infection, a known risk
group-A was 7.6 ± 3.2 years. Patients in group-A had factor for HCC. Cirrhosis itself is diabetogenic state and
more advanced HCC (size of lesion > 5 cm and >3 le- also a predisposition for HCC. Diabetes is a risk factor, if
sions of 3 cm or more diameter, portal vein thrombosis not actually etiologic for HCC, but temporal relationship
or intra-hepatic bile duct involvement) than group-B [34 is not yet clearly defined.9,27 Only few studies have shown
(73.9%) vs 72 (54.3%)]. Mortality with in one year was DM to precede HCC.1,38 World-wide, incidence of DM as
significantly more in group-A compared to group-B [36 well as HCC is increasing.1 India is experiencing an epi-
(78.2%) vs 56 (49.1%)]. Conclusion: DM is associated demic of DM.39 Establishing epidemiological relation and
with more advanced lesion and poor outcome in patient cause-effect relationship between these two is important.
with HCC. There are only few studies on impact of DM on man-
agement and outcome of HCC. None of the studies from
Head. India have shown relation of DM with HCC. This study
Clinical assistants, Gastroenterology Department, Bombay was planned to evaluate impact of DM on management
Hospital and Medical Research Centre, Mumbai. and outcome of HCC.
Address for correspondence:
Dr. Deepak Amarapurkar Materials and methods
D 401/402 Ameya RBI Employees
Co-Op Housing Society, This case-control observational study was carried out
Plot No. 947-950
New Prabhadevi Road
over study period of 10 years (1997-2006) on all the con-
Prabhadevi secutive patients of HCC. Diagnosis of HCC was based on
Mumbai 400 025 hyper-vascular tumor in the liver on two imaging studies
Telephone No. 91 22 24306262/24222432 or hyper-vascular tumor on single imaging modality with
Fax No. 91 22 24368623
E-mail: email@example.com; firstname.lastname@example.org
serum alpha-fetoprotein level greater than 400 ng/dL.
Patients were divided into 2 groups: a) Diabetic pa-
Manuscript received and accepted: 25 April 2008 tients with HCC and b) Non-diabetic patients with HCC.
DN Amarapurkar et al. Impact of diabetes mellitus on outcome of HCC 149
Diabetes was diagnosed according to American Diabetes Discussion
Association criteria on the basis of use of oral hypogly-
cemic drugs; fasting plasma glucose level ≥ 126 mg/dL; Previously, DM is shown to be a bad prognostic factor
2-hour plasma glucose ≥ 200 mg/dL during oral glucose for long-term survival of cirrhotic patients and mortality
tolerance test; and/or random or 2-hour post-prandial mainly being related to liver failure.40
plasma glucose level ≥ 200 mg/dL. Among 7 studies done to find out impact of DM on
In both groups following features were noted: age of HCC management, with few exceptions, most have
presentation, clinical features, laboratory features, Child shown increased post-resection complications and de-
class (CPT score), etiology (hepatitis B, hepatitis C, alco- creased post-resection survival, most probably due to in-
hol, NASH or other etiologies), tumor characteristics at creased risk of hepatic decompensation in DM.41-47 In our
presentation (advanced HCC), management (resection study, DM is associated with morphologically advanced
and ablative therapy or palliative management) and sur- lesions and with advanced liver disease regardless of eti-
vival. Advanced HCC (morphological) was diagnosed on ology. This can be a deciding factor for delineating man-
basis of tumor size > 5 cm or > 3 tumors each measuring > agement strategies, namely surgical resection, percutane-
3 cm diameter or portal vein thrombosis or bile duct in- ous therapies and trans-arterial chemoembolization. Our
vasion. study also shows that DM has adverse prognosis in HCC
Statistical analysis was performed using Chi square with high 1-year mortality rate.
test and student t test. It is still unclear whether HCC occurs because of insu-
lin resistance, which leads to NASH, which leads to cir-
Results rhosis, or whether the stimulatory effects of insulin on
hepatocyte growth lead more directly to neoplasia. Re-
As seen in Table I, majority of the patients in both cent studies have thrown light on how DM leads to HCC.
groups were in age group 51-70 years. There was no sta- DM is a state of hyperinsulinemia. Hyperinsulinemia
tistically significant difference in both the study groups may directly induce HCC: 1. by the up-regulation of re-
regarding age distribution. ceptors of specific growth factors (insulin and insulin-
Different etiologies of HCC are tabulated in Table II. like-growth factor-1);48,49 2. by activating mitogen acti-
As an etiology for HCC, alcohol and NASH were signifi- vated kinase that leads to phosphorylation of insulin re-
cantly higher in group-A. ceptor substance-1(IRS-1) a key protein involve in
As seen in Table III, in group-A there was higher child cellular proliferation.50,51 Insulin resistance may play a
class C patients and more advanced HCC. Also there was role by increasing oxidative stress and generation of reac-
lower rate of curative treatment for HCC. Mortality rates tive oxygen species that leads to a. p53 tumor suppressor
at 1-year were higher in group-A. gene mutation via by-product of lipid peroxidation (4-
hydroxynoneal),52,53 or b. up-regulation of proinflamma-
tory cytokines. Thus, inflammation, cellular prolifera-
tion, apoptosis inhibition and tumor suppressor gene mu-
Table I. Age-wise distribution. tations in setting of advanced liver disease (as a result of
insulin resistance and hyperinsulinemia) may lead to
Age in years, Group A, Group B,
n (%) n = 46 n = 114 p
< 50 8 (17.4) 17 (14.9) NS Table III. Presenting features and outcome of both groups.
51-60 13 (28.3) 28 (24.6) NS
61-70 19 (41.3) 46 (40.3) NS Group A, Group B,
> 70 6 (13) 23 (20.2) NS Parameters n = 46 n=114 p
Mean age, year 62.6 ± 9.5 66.7 ± 13.7 NS
Sex ratio (M:F) 6.6:1 5.4:1 NS
Mean duration of
Table II. Different etiologies of HCC. Diabetes, years 7.6 ± 3.2 --- -
Mean CPT score 11.8 ± 2.6 10.2 ± 2.1 NS
Etiology, n (%) Group A, Group B, Total Child class C, n (%) 37 (80.4) 70 (61.4) S
n = 46 n = 114 p n = 160 α-fetoprotein
(> 400 ng/dL), n (%) 16 (34.8) 46 (40.3) NS
Hepatitis B alone 11 (23.9) 34 (29.8) NS 45 (28.1) Mean α-fetoprotein,
Hepatitis B with alcohol 6 (13) 6 (5.3) NS 12 (7.5) ng/dL 1864.5 ± 441.5 2309 ± 726.6 NS
Alcohol alone 14 (30.4) 13 (11.4) S 27 (16.8) Advanced lesions, n (%) 34 (73.9) 62 (54.3) S
Hepatitis C alone 2 (4.3) 16 (14) NS 18 (11.3) Resection or ablative
Hepatitis C with alcohol 0 (0) 1 (0.9) NS 1 (0.6) therapy, n (%) 4 (8.6) 20 (17.5) S
NASH 4 (8.7) 0 (0) S 4 (2.5) Mean survival, months 10.1 ± 3.1 18.7 ± 6.1 S
Cryptogenic 9 (19.6) 44 (38.6) S 53 (33.1) Mortality at 1-year, n (%) 36 (78.2) 56 (49.1) S
150 Annals of Hepatology 7(2) 2008: 148-151
HCC formation. Hyperinsulinemia is associated with in- 21. Wideroff L, Gridley G, Mellemkjaer L, Chow WH, Linet M,
creased risk of HCC as well as it is shown to increase Keehn S, et al. Cancer incidence in a population-based cohort of
patients hospitalized with diabetes mellitus in Denmark. J Natl
growth rate of HCC.54,55 Cancer Inst 1997; 89: 1360-1365.
In conclusion, Diabetic patients with HCC had more 22. Lagiou P, Kuper H, Stuver S, Tzonou A, Trichopoulos D, Adami
advanced liver failure, morphologically more advanced HO. Role of diabetes mellitus in the etiology of hepatocellular
lesions and poorer long-term prognosis compared to non- carcinoma. J Natl Cancer Inst 2000; 92: 1096-1099.
23. Fujino Y, Mizoue T, Tokui N, Yoshimura T. Prospective study
diabetic patients with HCC. of diabetes mellitus and liver cancer in Japan. Diabetes Metab
Res Rev 2001; 17: 374-379.
References 24. Tazawa J, Maeda M, Nakagawa M, Ohbayashi H, Kusano F,
Yamane M, et al. Diabetes mellitus may be associated with
hepatocarcinogenesis in patients with chronic hepatitis C. Dig Dis
1. El-Serag HB, Tran T, Everhart JE. Diabetes increases the risk of
Sci 2002; 47: 710-715.
chronic liver disease and hepatocellular carcinoma. Gastroenter-
25. Hassan M, Hwang L, Hatten C, Swaim M, Li D, Abbruzzese J, et
ology 2004; 126: 460-468.
al. Risk factors for hepatocellular carcinoma: synergism of alco-
2. Amarapurkar D, Das HS. Chronic liver disease in diabetes melli-
hol with viral hepatitis and diabetes mellitus. Hepatology 2002;
tus. Tropical Gastroenterol 2002; 23: 3-5.
3. Ruhl CE, Everhart JE. Determinants of the association of over-
26. Lai MS, Hsieh MS, Chiu YH, Chen TH. Type 2 diabetes and
weight with elevated serum alanine aminotransferase activity in
the United States. Gastroenterology 2003; 124: 71-79. hepatocellular carcinoma: a cohort study in high prevalence area
4. Di Bisceglie AM, Carithers Jr RL, Gores GL. Hepatocellular car- of hepatitis virus infection. Hepatology 2006; 43: 1295-1302.
cinoma. Hepatology 1998; 28: 1161-1165. 27. Harrison SA. Liver disease in patients with diabetes mellitus. J
5. El-Serag HB, Richardson PA, Everhart JE. The role of diabetes in Clin Gastroenterol 2006; 40: 68-76.
hepatocellular carcinoma: a case-control study among United 28. Kaczynski J, Hansson G, Wallerstedt S. Diabetes: one of few
States veterans. Am J Gastroenterol 2001; 96: 2462-2467. remarkable differences in clinicopathologic features between cir-
6. Kessler II. Cancer mortality among diabetics. J Natl Cancer Inst rhotic and noncirrhotic Swedes with hepatocellular carcinoma.
1970; 44: 2051-2055. Dig Dis Sci 2006; 51: 796-802.
7. Ragozzino M, Melton III LJ, Chu CP, Palumbo PJ. Subsequent 29. Di Bisceglie AM. What every hepatologist should know about
cancer risk in the incidence cohort of Rochester, Minnesota, resi- endocrinology: Obesity, diabetes, and liver disease. Gastroenter-
dents with diabetes mellitus. J Chronic Dis 1982; 35: 13-19. ology 2004; 126: 604-606.
8. Lu SN, Lin TM, Chen CJ, Chen JS, Liaw YF, Chang WY, et al. A 30. Matteoni C, Younossi Z, Gramlich T, Bopari N, Liu Y,
case-control study of primary hepatocellular carcinoma in Tai- McCullough A. Nonalcoholic fatty liver disease: a spectrum of
wan. Cancer 1988; 62: 2051-2055. clinical and pathological severity. Gastroenterology 1999; 116:
9. Beasley RP. Diabetes and hepatocellular carcinoma. Hepatology 1413-1419.
2006; 44: 1408-1410. 31. Marchesini G, Brizi M, Morselli-Labate A, Bianchi G, Bugianesi
10. El–Serag HB, Hampel H, Javadi F. The Association Between E, McCullough A, et al. Association of nonalcoholic fatty liver
Diabetes and Hepatocellular Carcinoma: A Systematic Review of disease with insulin resistance. Am J Med 1999; 107: 450-455.
Epidemiologic Evidence. Clin Gastroenterol Hepatol 2006; 4: 32. Belfiore F, Iannello S. Insulin resistance in obesity: metabolic
369-380. mechanisms and measurement methods. Mol Genet Metab 1998;
11. La Vecchia C, Negri E, D’Avanzo B, Boyle P, Franceschi SD. 65: 121-128.
Medical history and primary liver cancer. Cancer Res 1990; 50: 33. Falck-Ytter, Younossi Z, Marchesini G, McCullough A. Clinical
6274-6277. features and natural history of nonalcoholic steatosis syndromes.
12. La Vecchia C, Negri E, Franceschi SD, D’Avanzo B, Boyle P. A Semin Liver Dis 2001; 21: 17-26.
case-control study of diabetes mellitus and cancer risk. Br J Can- 34. Powell E, Cooksley W, Hanson R, Searle J, Halliday J, Powell L.
cer 1994; 70: 950-953. The natural history of nonalcoholic steatohepatitis: a follow-up
13. La Vecchia C, Negri E, DeCarli A, Franceschi SD. Diabetes mel- study of forty-two patients for up to 21 years. Hepatology 1990;
litus and the risk of primary liver cancer. Int J Cancer 1997; 73: 11: 74-80.
204-207. 35. Cotrim H, Parana R, Braga E, Lyra L. Nonalcoholic steatohepatitis
14. Adami HO, McLaughlin J, Ekbom A, Berne C, Silverman D, and hepatocellular carcinoma: natural history? Am J Gastroenterol
Hacker D, et al. Cancer risk in patients with diabetes mellitus. 2000; 95: 3018-3019.
Cancer Causes Control 1991; 2: 307-314. 36. Zen Y, Katayanagi K, Tsuneyama K, Harada K, Araki I,
15. Adami HO, Chow WH, Nyren O, Berne C, Linet MS, Ekbom A, Nakanuma Y. Hepatocellular carcinoma arising in non-alcoholic
et al. Excess risk of primary liver cancer in patients with diabetes steatohepatitis. Pathol Int 2001; 51: 127-131.
mellitus. J Natl Cancer Inst 1996; 88: 1472-1477. 37. Shimada M, Hashimoto E, Taniai M, Hasegawa K, Okuda H,
16. Kingston ME, Ali MA, Atiyeh M, Donnelly RJ. Diabetes mellitus Yayashi N, et al. Hepatocellular carcinoma in patients with non-
in chronic active hepatitis and cirrhosis. Gastroenterology 1984; alcoholic steatohepatitis. J Hepatol 2002; 37: 154-160.
87: 688-694. 38. Harris MI, Klein R, Welborn TA, Knuiman MW. Onset of NIDDM
17. Lawson DH, Gray JM, McKillop C, Clarke J, Lee FD, Patrick RS. occurs at least 4–7 years before clinical diagnosis. Diabetes Care
Diabetes mellitus and primary hepatocellular carcinoma. Q J Med 1992; 15: 815-819.
1986; 61: 945-955. 39. Iyer SR. Type 2 diabetes mellitus express highway, where is the
18. Davila JA, Morgan RO, Shaib Y, et al. Diabetes increases the risk ‘U’ turn? JAPI 2003; 51: 495-500.
of hepatocellular carcinoma in the United States: a population 40. Bianchi G, Marchesini G, Zoli M, Bugianesi E, Fabbri A, Pisi E.
based case control study. Gut 2005; 54: 533-539. Prognostic significance of diabetes in patients with cirrhosis.
19. Yu MC, Tong MJ, Govindarajan S, Henderson BE. Non-viral Hepatology 1994; 20: 119-25.
risk factors for hepatocellular carcinoma in a low-risk popula- 41. Huo T-I, Lui W-Y, Huang Y-H, et al. Diabetes mellitus is a risk
tion, the non-Asians of Los Angeles County, California. J Natl factor for hepatic decompensation in patients with hepatocellular
Cancer Inst 1991; 83: 1820-1826. carcinoma undergoing resection: a longitudinal study. Am J
20. Braga C, La Vecchia C, Negri E, Franceschi S. Attributable risks Gastroenterol 2003; 98: 2293-2298.
for hepatocellular carcinoma in Northern Italy. Eur J Cancer 42. Huo T-I, Wu J-C, Lui W-Y, et al. Differential mechanism and
1997; 33: 629-634. prognostic impact of diabetes mellitus on patients with hepato-
DN Amarapurkar et al. Impact of diabetes mellitus on outcome of HCC 151
cellular carcinoma undergoing surgical and nonsurgical treat- vation by hepatitis B virus X gene products. Cancer Res 1996;
ment. Am J Gastroenterol 2004; 99: 1479-1487. 56: 3831-36.
43. Toyoda H, Kumada T, Nakano S, et al. Impact of diabetes melli- 50. Kaburagi Y, Yamachui T, Yamamoto-Honda R, et al. The mecha-
tus on the prognosis of patients with hepatocellular carcinoma. nism of insulin-induced signal transduction mediated by the insu-
Cancer 2001; 91: 957-963. lin receptor substrate family. Endocr J 1999; 56(suppl): 25-34.
44. Ikeda Y, Shimada M, Hasegawa H, et al. Prognosis of hepatocel- 51. Tanaka S, Mohr L, Schmidt EV, et al. Biologic effect of human
lular carcinoma with diabetes mellitus after hepatic resection. insulin receptor substrat-1 overexpression in hepatocytes.
Hepatology 1998; 27: 1567-1571. Hepatology 1997; 26: 598-604.
45. Poon RT, Fan ST, Wong J. Does diabetes mellitus influence the 52. Hu W, Feng Z, Eveleigh J, et al. The major lipid peroxidation
perioperative outcome or long-term prognosis after resection of product, trans-4-hydroxy-2-noneal, preferentially from DNA
hepatocellular carcinoma? Am J Gastroenterol 2002; 97: 1480-1488. adduct at codon 249 of human p53gene, a unique mutational
46. Shimada M, Matsumata T, Akazawa K, et al. Estimation of risk of hotspot in hepatocellular carcinoma. Carcinogenesis 2002; 23:
major complications after hepatic resections. Am J Surg 1994; 1781-89.
167: 339-403. 53. Hsu HC, Peng SY, Lai PL, et al. Allotype loss of heterogeneity of
47. Yanaga K, Matsumata T, Hayashi H, et al. Effect of diabetes p53 in primary and recurrent hepatocellular carcinoma: a stuffy
mellitus on hepatic resection. Arch Surg 1993; 128: 445-448. of 150 patient. Cancer 1994; 73: 42-7.
48. Moore MA, Park CB, Tsuda H. Implication of the 54. Balkau B, Kahn HS, Courbon D, et al. Hyperinsulinemia predicts
hyperinsulinemia-diabetes-cancer link for preventive effect. Eur fatal liver cancer but is inversely associated with fatal cancer at
J Cancer Prev 1998; 7: 89-107. some other sites. Diabetes Care 2001; 24: 843-849.
49. Kim SO, Park GJ, Lee YI. Increased expression of the insulin like 55. Saito K, Inoue S, Saito T, et al. Augmentation effect of postpran-
growth factor receptor –I (IGF-1) receptor gene in hepatocellu- dial hyperinsulinemia on growth of human hepatocellular carci-
lar carcinoma cell lines: implication of IGF-I receptor gene acti- noma. Gut 2002; 51: 100-104.